Nonstationary Gaussian Process Discriminant Analysis With Variable Selection for High-Dimensional Functional Data

High-dimensional classification and feature selection tasks are ubiquitous with the recent advancement in data acquisition technology. In several application areas such as biology, genomics, and proteomics, the data are often functional in their nature and exhibit a degree of roughness and nonstationarity. These structures pose additional challenges to commonly used methods that rely mainly on a two-stage approach performing variable selection and classification separately. We propose in this work a novel Gaussian process discriminant analysis (GPDA) that combines these steps in a unified framework. Our model is a two-layer nonstationary Gaussian process coupled with an Ising prior to identify differentially-distributed locations. Scalable inference is achieved via developing a variational scheme that exploits advances in the use of sparse inverse covariance matrices. We demonstrate the performance of our methodology on simulated datasets and two proteomics datasets: breast cancer and SARS-CoV-2. Our approach distinguishes itself by offering explainability as well as uncertainty quantification in addition to low computational cost, which are crucial to increase trust and social acceptance of data-driven tools. Supplementary materials for this article are available online.

Future preferred mode of learning of business undergraduates and its implications

e-Learning was abruptly adopted in many countries to mitigate the adverse consequences of the sudden closure of institutions of higher learning caused by the COVID-19 pandemic. Against this background, this study investigated how business undergraduates want to learn in the future and predictors of their future preferred mode of learning. 251 business undergraduates from a private university in Malaysia participated in an online survey conducted in July 2020, during the sudden closure of institutions of higher learning. Data collected were analysed using the multiple discriminant analysis to develop a characteristics profile of the three groups of business undergraduates (i.e., preferred fully conventional classroom learning, blended learning and fully e-learning) in terms of important predictors. Results revealed that the significant predictors of future preferred mode of learning of business undergraduates, in descending order, were disadvantages of e-learning, advantages of e-learning, self-regulated learning, learning outcomes, information and communications technology infrastructure and training, support and resources. This study concludes with some reflective thoughts about important lessons learned from this unprecedented pandemic pertaining to e-learning readiness to deal with future unexpected crises.

EMPLOYING ARTIFICIAL NEURAL NETWORKS AND MULTIPLE DISCRIMINANT ANALYSIS TO EVALUATE THE IMPACT OF THE COVID-19 PANDEMIC ON THE FINANCIAL STATUS OF JORDANIAN COMPANIES

Aim/Purpose This paper aims to empirically quantify the financial distress caused by the COVID-19 pandemic on companies listed on Amman Stock Exchange (ASE). The paper also aims to identify the most important predictors of financial distress pre- and mid-pandemic. Background The COVID-19 pandemic has had a huge toll, not only on human lives but also on many businesses. This provided the impetus to assess the impact of the pandemic on the financial status of Jordanian companies. Methodology The initial sample comprised 165 companies, which was cleansed and reduced to 84 companies as per data availability. Financial data pertaining to the 84 companies were collected over a two-year period, 2019 and 2020, to empirically quantify the impact of the pandemic on companies in the dataset. Two approaches were employed. The first approach involved using Multiple Discriminant Analysis (MDA) based on Altman's (1968) model to obtain the Z-score of each company over the investigation period. The second approach involved developing models using Artificial Neural Networks (ANNs) with 15 standard financial ratios to find out the most important variables in predicting financial distress and create an accurate Financial Distress Prediction (FDP) model. Contribution This research contributes by providing a better understanding of how financial distress predictors perform during dynamic and risky times. The research confirmed that in spite of the negative impact of COVID-19 on the financial health of companies, the main predictors of financial distress remained relatively steadfast. This indicates that standard financial distress predictors can be regarded as being impervious to extraneous financial and/or health calamities. Findings Results using MDA indicated that more than 63% of companies in the dataset have a lower Z-score in 2020 when compared to 2019. There was also an 8% increase in distressed companies in 2020, and around 6% of companies came to be no longer healthy. As for the models built using ANNs, results show that the most important variable in predicting financial distress is the Return on Capital. The predictive accuracy for the 2019 and 2020 models measured using the area under the Receiver Operating Characteristic (ROC) graph was 87.5% and 97.6%, respectively. Recommendations Decision makers and top management are encouraged to focus on the identified for Practitioners highly liquid ratios to make thoughtful decisions and initiate preemptive actions to avoid organizational failure. Recommendations This research can be considered a stepping stone to investigating the impact of for Researchers COVID-19 on the financial status of companies. Researchers are recommended to replicate the methods used in this research across various business sectors to understand the financial dynamics of companies during uncertain times. Impact on Society Stakeholders in Jordanian-listed companies should concentrate on the list of most important predictors of financial distress as presented in this study. Future Research Future research may focus on expanding the scope of this study by including other geographical locations to check for the generalisability of the results. Future research may also include post-COVID-19 data to check for changes in results. © 2023 Informing Science Institute. All rights reserved.

Major Depressive Disorder and Chronic Fatigue Syndrome Show Characteristic Heart Rate Variability Profiles Reflecting Autonomic Dysregulations: Differentiation by Linear Discriminant Analysis.

Major depressive disorder (MDD) and chronic fatigue syndrome (CFS) have overlapping symptoms, and differentiation is important to administer the proper treatment. The present study aimed to assess the usefulness of heart rate variability (HRV) indices. Frequency-domain HRV indices, including high-frequency (HF) and low-frequency (LF) components, their sum (LF+HF), and their ratio (LF/HF), were measured in a three-behavioral-state paradigm composed of initial rest (Rest), task load (Task), and post-task rest (After) periods to examine autonomic regulation. It was found that HF was low at Rest in both disorders, but was lower in MDD than in CFS. LF and LF+HF at Rest were low only in MDD. Attenuated responses of LF, HF, LF+HF, and LF/HF to task load and an excessive increase in HF at After were found in both disorders. The results indicate that an overall HRV reduction at Rest may support a diagnosis of MDD. HF reduction was found in CFS, but with a lesser severity. Response disturbances of HRV to Task were observed in both disorders, and would suggest the presence of CFS when the baseline HRV has not been reduced. Linear discriminant analysis using HRV indices was able to differentiate MDD from CFS, with a sensitivity and specificity of 91.8% and 100%, respectively. HRV indices in MDD and CFS show both common and different profiles, and can be useful for the differential diagnosis.

Preliminary Diagnosis of COVID-19 using Speech Processing Techniques

This project aims to devise an alternative for Coronavirus detection using various audio signals. The aim is to create a machine-learning model assisted by speech processing techniques that can be trained to distinguish symptomatic and asymptomatic Coronavirus cases. Here the features exclusive to the vocal cord of a person is used for covid detection. The procedure is to train the classifier using a data set containing data of people of various ages both infected and disease-free, including patients with comorbidities. We presented a machine learning-based Coronavirus classifier model that can separate Coronavirus positive or negative patients from cough, breathing, and speech recordings. The model was trained and evaluated using several machine learning classifiers such as Random Forest Classifier, Logistic Regression (LR), Decision Tree Classifier, k-nearest Neighbour (KNN), Naive Bayes Classifier, Linear Discriminant Analysis, and a neural network. This project helps track COVID-19 patients at a low cost using a non-contactable procedure and reduces the workload on testing centers. © 2023 IEEE.

A LASSO-based model for financial distress of the Vietnamese listed firms: Does the covid-19 pandemic matter?

Financial distress is a vexing managerial challenge for businesses worldwide, especially during a turbulent period like the COVID-19 pandemic. Motivated by an increasing number of closed businesses in Vietnam during the recent COVID-19 pandemic, this study is conducted to provide a comprehensive analysis of financial distress for Vietnamese listed firms. Machine learning approaches are employed using the annual data of 492 listed firms from 2012 to 2021. Specifically, we aim to identify the appropriate distress predictors for the Vietnamese listed firms using LASSO, a technique known to be superior compared to other variable selection techniques. Empirical results reveal that there are four key financial distress predictors for the Vietnamese listed firms, namely the ratios of (i) working capital and total assets, (ii) retained earnings and total assets, (iii) earnings before interest and taxes and total assets and (iv) net income and total assets. We also conducted an industry-level analysis and found that the Energy sector experienced the highest number of financially distressed firms during Covid-19. In contrast, Communication Services, Health Care, and Utilities had the lowest number of distressed firms. Policy implications have emerged based on these important findings from our analysis.

Analysis from computer science and systemic analysis of the discriminant analysis for the rural working population of the canton Milagro

Rural coexistence for many people is associated with living in a constant disadvantage, which has increased with the emergence of the pandemic generated by covid-19, for this reason we seek to give an interpretation to the behavior presented by the study group. For the purposes of the research work, the multivariate statistical technique of Discriminant Analysis was used, applied to a group of variables in two surveys conducted to those who make up the working population, emphasizing that the questionnaires refer to two different periods of time, where it was found that for reasons of the scenario in which they perform their activities, the vast majority fails to have the right lifestyle, The same gains more value from comments made by the population, where they highlight aspects such as the high crime rate, the low degree of professionalization, abuse by third parties, and the growing difficulty to mitigate this social crisis given the lack of labor supply, because the employed people fulfill their working hours, but do not enjoy a fixed working day with a constant and adequate remuneration, in addition to the benefits that a job in a dependent relationship could give them. © 2022, Associacao Iberica de Sistemas e Tecnologias de Informacao. All rights reserved.

Classification of Respiratory Syncytial Virus and Sendai Virus Using Portable Near-Infrared Spectroscopy and Chemometrics

There is evidence that it may be possible to detect viruses and viral infection optically using techniques such as Raman and infrared (IR) spectroscopy and hence open the possibility of rapid identification of infected patients. However, high-resolution Raman and IR spectroscopy instruments are laboratory-based and require skilled operators. The use of low-cost portable or field-deployable instruments employing similar optical approaches would be highly advantageous. In this work, we use chemometrics applied to low-resolution near-IR (NIR) reflectance/absorbance spectra to investigate the potential for simple low-cost virus detection suitable for widespread societal deployment. We present the combination of near-IR spectroscopy (NIRS) and chemometrics to distinguish two respiratory viruses, respiratory syncytial virus (RSV), the principal cause of severe lower respiratory tract infections in infants worldwide, and Sendai virus (SeV), a prototypic paramyxovirus. Using a low-cost and portable spectrometer, three sets of RSV and SeV spectra, dispersed in phosphate-buffered saline (PBS) medium or Dulbecco's modified eagle medium (DMEM), were collected in long- and short-term experiments. The spectra were preprocessed and analyzed by partial least-squares discriminant analysis (PLS-DA) for virus type and concentration classification. Moreover, the virus type/concentration separability was visualized in a low-dimensional space through data projection. The highest virus-type classification accuracy obtained in PBS and DMEM is 85.8% and 99.7%, respectively. The results demonstrate the feasibility of using portable NIR spectroscopy as a valuable tool for rapid, on- site, and low-cost virus prescreening for RSV and SeV with the further possibility of extending this to other respiratory viruses such as SARS-CoV-2.

Authentication of Tinospora cordifolia derived herbal supplements using high resolution mass spectrometry-based metabolomics approach – A pilot study

Tinospora cordifolia herbal supplements have recently gained prominence due to their promising immunomodulatory and anti-viral effects against SARS-CoV-2. Mislabelling or diluting Tinospora supplements for profit may harm public health. Thus, validating the label claim of these supplements in markets is critical. This study investigated how high resolution mass spectrometry-based metabolomics and chemometrics can be used to distinguish Tinospora cordifolia from two other closely related species (T. crispa and T. sinensis). The Orthogonal Partial Least Square Discriminant Analysis (OPLS-DA) and PLS-DA based chemometric models predicted the species identity of Tinospora with 94.44% accuracy. These classification models were trained using 54 T. cordifolia, 21 T. crispa, and 21 T. sinensis samples. We identified 7 biomarkers, including corydine, malabarolide, ecdysterone, and reticuline, which discriminated Tinospora cordifolia from the two other species. The label claim of 25 commercial Tinospora samples collected from different parts of India was verified based on the relative abundance of the biomarker compounds, of which 20 were found authentic. The relative abundance of biomarkers significantly varied in the 5 suspicious market samples. This pilot study demonstrates a robust metabolomic approach for authenticating Tinospora species, which can further be used in other herbal matrices for product authentication and securing quality. © 2023 Elsevier B.V.

Evaluation of Hand Washing Procedure Using Vision-Based Frame Level and Spatio-Temporal Level Data Models

Hand hygiene is obligatory for all healthcare workers and vital for patient care. During COVID-19, adequate hand washing was among recommended measures for preventing virus transmission. A general hand-washing procedure consisting several steps is recommended by World Health Organization for ensuring hand hygiene. This process can vary from person to person and human supervision for inspection would be impractical. In this study, we propose computer vision-based new methods using 12 different neural network models and 4 different data models (RGB, Point Cloud, Point Gesture Map, Projection) for the classification of 8 universally accepted hand-washing steps. These methods can also perform well under situations where the order of steps is not observed or the duration of steps are varied. Using a custom dataset, we achieved 100% accuracy with one of the models, and 94.23% average accuracy for all models. We also developed a real-time robust data acquisition technique where RGB and depth streams from Kinect 2.0 camera were utilized. Results showed that with the proposed methods and data models, efficient hand hygiene control is possible.

Assessment of the financial condition of selected dairy cooperatives during the COVID-19 pandemic

The aim of the article is to present the financial condition of selected dairy cooperatives using ratio analysis and selected discriminant models. The main objective of the paper is to assess the overall financial condition of dairy cooperatives during the COVID-19 pandemic (2020-2021) and earlier years (2017-2019). The author focused, on the one hand, on the assessment of the financial condition of a selected group and, on the other hand, on the link between the financial situation of selected dairy cooperatives and state aid during the changing economic reality caused by the SARS CoV-2 virus. The financial analysis for dairy cooperatives also reveals a broader comparative context in the time span before and during the COVID-19 pandemic. The research shows that the analysed dairy cooperatives, with the exception of OSM Jasienica Rosielna, did not have a negative financial results.

Automated COVID-19 Detection from WBC-DIFF Scattergram Images with Hybrid CNN Model Using Feature Selection

In the medical diagnosis such as WBC (white blood cell), the scattergram images show the relationships between neutrophils, eosinophils, basophils, lymphocytes, and monocytes cells in the blood. For COVID-19 detection, the distributions of these cells differ in healthy and COVID-19 patients. This study proposes a hybrid CNN model for COVID-19 detection using scatter images obtained from WBC sub (differential-DIFF) parameters instead of CT or X-Ray scans. As a data set, the scattergram images of 335 COVID-19 suspects without chronic disease, collected from the biochemistry department of Elazig Fethi Sekin City Hospital, are examined. At first, the data augmentation is performed by applying HSV(Hue, Saturation, Value) and CIE-1931(Commission Internationale de l'éclairage) conversions. Thus, three different image large sets are obtained as a result of raw, CIE-1931, and HSV conversions. Secondly, feature extraction is applied by giving these images as separate inputs to the CNN model. Finally, the ReliefF feature extraction algorithm is applied to determine the most dominant features in feature vectors and to determine the features that maximize classification accuracy. The obtaining feature vector is classified with high-performance SVM in binary classification. The overall accuracy is 95.2%, and the F1-Score is 94.1%. The results show that the method can successfully detect COVID -19 disease using scattergram images and is an alternative to CT and X-Ray scans.

A New Respiratory Diseases Detection Model in Chest X-Ray Images Using CNN

Convolutional Neural Network (CNN)-based deep learning techniques have recently demonstrated increased potential and effectiveness in image recognition applications, such as those involving medical images. Deep-learning models can recognize targets with performance comparable to radiologists when used with CXR. The primary goal of this research is to examine a deep learning technique used on the radiography dataset to detect COVID-19 in X-ray medical images. The proposed system consists of several stages, from pre-processing, passing through the feature reduction using more than one technique, to the classification stage based on a proposed model. The test was applied to the COVID-19 Radiography dataset of normal and three lung infections (COVID-19, Viral Pneumonia, and Lung Opacity). The proposed CNN model has shown its ability to classify COVID, normal, and other lung infections with perfect accuracy of 99.94%. Consequently, the AI-based early-stage detection algorithms will be enhanced, increasing the accuracy of the X-raybased modality for the screening of various lung diseases. © 2023 Lavoisier. All rights reserved.

Personality and COVID-19 vaccine hesitancy: A quantitative study

Despite rigorous communication and education programs, hesitancy rates for the COVID-19 vaccination in the U.S. remain relatively unchanged at 20-25% (Kaiser Family Foundation, September 2022). Increasing the vaccination uptake to prevent further spread of the disease requires a better understanding of underlying psychological reasons for vaccine hesitancy. This study examines whether any Big Five personality traits (extroversion, agreeableness, conscientiousness, openness, and emotional stability) could statistically significantly predict group membership into one of four vaccination status groups of adults living in the U.S. The four vaccination groups were: (Group 1) did not have the COVID-19 vaccine and will not in the future;(Group 2) did not have the vaccine but may in the future;(Group 3) had the vaccine but did not intend to have another COVID vaccine in the future, and (Group 4) had the vaccine and intended on getting future vaccinations as appropriate. Discriminant analysis was used as a statistical test to predict group membership into one of the four vaccination groups from the mean scores of a brief personality inventory. Adult participants from across the country were recruited using CloudResearch Managed Survey (N = 119) using a survey uploaded from Qualtrics. The respondents took a brief online psychological assessment, the Ten Item Personality Inventory (TIPI) (Gosling, 2003), and were asked about their current vaccine status and future intention to vaccinate. Demographic questions regarding political affiliation and religiosity were also included to understand the population better and add value to the analysis. This study did not support the hypothesis that personality type, measured by the TIPI, would statistically significantly predict membership into vaccine status/ intention groups. (PsycInfo Database Record (c) 2023 APA, all rights reserved)

Psychological Profiles of Positive Mental Health of Women in Situation of Confinement Due to COVID-19

Background: The world health organization has indicated that the problem of COVID-19 and confinement generated strong psychological impacts on the world population. Much of the research has focused on studying mental health in different population groups, leaving aside a positive mental health perspective. Objective(s): The present research intended to establish positive mental health profiles in confined women due to COVID-19 during 2020. Method(s): The factors of personal satisfaction, prosocial attitude, self-control, autonomy, problem-solving, self-actualization, and interpersonal relationships were assessed through the application of Lluch's positive mental health scale, in a sample of 202 confined women in the Department of Sucre, Colombia, selected by non-probabilistic convenience sampling. In addition, cluster analysis models were applied to identify psychological profiles of positive mental health and characterize sociodemographic variables, the selected model was evaluated and validated using the statistical technique of discriminant analysis using Minitab 18 software. Results and Discussion: A positive mental health differentiation in women is shown from which four psychological profiles of positive mental health could be identified, with scores of 14.10 in profile 1, 11.41 in profile 2, 9.15 in profile 3, and 7.56 in profile 4. The positive mental health factors used showed an ability to discriminate in 92.6% of the cases in the profiles. Conclusion(s): The identified profiles are significant and important to characterize psychometric profiles of positive mental health of confined women, which are important results for their diagnosis and the development of public policies for their treatment.Copyright © 2023 Bentham Science Publishers.

Leveraging Twitter data to analyze the virality of Covid-19 tweets: a text mining approach

As the novel coronavirus spreads across the world, work, pleasure, entertainment, social interactions, and meetings have shifted online. The conversations on social media have spiked, and given the uncertainties and new policies, COVID-19 remains the trending topic on all such platforms, including Twitter. This research explores the factors that affect COVID-19 content-sharing by Twitter users. The analysis was conducted using 57,000 plus tweets that mentioned COVID-19 and related keywords. The tweets were subjected to the Natural Language Processing (NLP) techniques like Topic modelling, Named Entity-Relationship, Emotion & Sentiment analysis, and Linguistic feature extraction. These methods generated features that could help explain the retweet count of the tweets. The results indicate that tweets with named entities (person, organisation, and location), expression of negative emotions (anger, disgust, fear, and sadness), reference to mental health, optimistic content, and greater length have higher chances of being shared (retweeted). On the other hand, tweets with more hashtags and user mentions are less likely to be shared.

Proteomic profiling of severe lung diffusion impairment in the recovery stage of COVID-19-induced ARDS

Background: Around 80% of patients who developed COVID-19-driven ARDS present lung ailment. There is a lack of knowledge of the mechanisms that mediate the pulmonary outcomes. Aim(s): To characterize the factors linked to diffusion impairment in survivors of severe COVID-19. Method(s): Prospective cohort study including 87 COVID-19-induced ARDS survivors. A complete pulmonary evaluation was performed 3 months after hospital discharge. 364 proteins were quantified using the proximity extension assay (PEA). Partial least square-discriminant analysis (PLS-DA) and random forest (RF) were used for multivariable analyses. Result(s): Moderate to severe diffusion impairment (DLCO<60% predicted) was observed in the 30% of the cohort. 15 proteins were differentially detected [false discovery rate (FDR)<0.05] in the univariate analysis. Pleiotrophin showed the highest differences (fold change=2.22 and FDR=0.001). In continuous analysis, proteins were inversely and independently associated with DLCO, and in some cases showed a robust dose-response relationship. PLS-DA and RF identified proteomic profiles related to the severity of diffusion capacity. Clusters identified were enriched in mediators of cell proliferation and differentiation, tissue remodeling, angiogenesis, coagulation, inflammation, immune response and fibrosis. Proteins are expressed in immune and non-immune lung cells. Conclusion(s): In survivors of COVID-19-driven ARDS, lung dysfunction is linked to plasma factors involved in injury and repair mechanisms. The host proteomic profile provides a novel understanding of post-acute sequelae and may be source of therapeutic strategies and biomarkers.

Companies Identified as Having the Greatest Returns to Capital in Emerging Markets During a Worldwide Pandemic

The COVID-19 pandemic of the year 2020 resulted in high unemployment, business closings, property loss and decimation of individual wealth, disruption of global supply chains, and illness and deaths everywhere, but most intensely in countries classified as emerging markets. However, during this year, cash flow from investors in established markets to emerging markets has been of immense magnitude. While many companies, in emerging markets, reported very high risk-adjusted rates of return, many others reported so low rates during this period. This study aims to establish a unique profile of risk-return characteristics of the companies in emerging markets that have constantly reported the highest risk-adjusted returns to total capital during the pandemic. The statistical results of our study suggest that such unique profile can be used as a tool to forecast which companies, in such markets and during such disturbances in the future, will maintain high returns to capital providing an invaluable tool for investors, investment counselors and financial researchers tasked to determine firm's intrinsic value in such an environment.

Comparison of Machine Learning Models for predicting Covid-19 patients' recovery in India

The pandemic Covid-19 is a name coined by WHO on 31st December 2019. This devastating illness was carried on by a new coronavirus known as SARS-COV-2. Most of the research has focused on estimating the total number of cases and mortality rate of COVID-19. Due to this, people across the world were stressed out by observing the growing number of cases every day. As a means of maintaining equilibrium, this paper aims to identify the best way to predict the number of recovered cases of Coronavirus in India. Dataset was divided into two parts: training and testing. The training dataset utilised 70% of the dataset, and the testing dataset utilised 30%. In this paper, we applied 10 machine learning techniques i.e. Random Forest Classifier (RF), Naive Bayes (NB), Quadratic Discriminant Analysis (QDA), Gradient Boosting Classifier (GBM), Linear Discriminant Analysis (LDA), Logistic Regression (LR), K Neighbour Classifier (KNN), Decision Tree Classifier (DT), SVM - Linear and Ada-Boost Classifier in order to predict recovered patients in India. Our study suggests that Random Forest Classifier outperforms other machine learning models for predicting the recovered Coronavirus patients having an accuracy of 0.9632, AUC of 0.9836, Recall of 0.9640, Precision of 0.9680, F1 Score of 0.9617 and Kappa of 0.9558. © 2022 IEEE.

Collaborative corporate social responsibility praxis: case studies from India

Purpose>This study aims to explore how corporate social responsibility (CSR) has assumed a new meaning today, with the COVID-19 pandemic. This, in turn, has changed the way companies now view the impact of their activities on the environment, customers, employees, community and other stakeholders.Design/methodology/approach>This paper uses a qualitative case study approach and draws a critical lens to document the complex interplay between dimensions of CSR, business sustainability and social issues, applying theoretical tools such as social capital theory and stakeholder theory to elucidate the nature of collaborative managerial responses to the organisation's challenges during the pandemic. This is a case study paper. This paper applies multi method approach to develop a case study analysis through participant observation and report analysis to investigate the CSR approaches undertaken in India by Infosys Genesis, a global leader in technology services and consulting, and Akshaya Patra Foundation, a non-governmental organisation (NGO), which operates the world's largest lunch school program. This was an appropriate methodology since the focus was on an area that was little understood, while the analysis required an in-depth understanding of a complex phenomenon through observation and a case study. In addition, case study research has been recommended for how, why and what type of research questions that focus on contemporary events (Saunders et al., 2003;Yin, 1994), such as CSR participation in the existing business environment. Furthermore, the issue under investigation is a real-life situation where the limitations between the phenomenon and the body of knowledge are unclear (Yin, 1994). This was the case because CSR has been probed by numerous disciplines through the application of various theoretical frameworks, each interpreting the context from their own perspective. Leximancer was used for the analysis (a text-mining software for visualising the structure of concepts and themes across case studies). This process differs from the traditional content analysis in that specific word strings are not needed;instead, Leximancer recognises what concepts are present in a set of texts, permitting concepts to be automatically coded in a grounded fashion (Cretchley et al., 2010, p. 2). The paper will be looked at from three levels comprising themes, concepts and concept profiling to create rich and reliable dimensions of a theoretical model (Myers, 2008). The themes are created in Leximancer software and are built on an algorithm that looks for hidden repeated patterns in interactions. The concepts add a layer and discover which concepts are shared by actors. The concept profiling allows to discover additional concepts and allows to do a discriminant analysis on prior concepts (Cretchley et al., 2010). Words that come up frequently are treated as concepts. Although the limited number of cases does not represent the entire sector, it enabled collection of rich data through quotes revealing some of the most crucial aspects of large organisations and non-profits in India.Findings>The findings demonstrate how these robust, innovative, collaborative CSR initiatives between a multinational firm and an NGO have been leveraged to combat manifold issues of education, employment and hunger during the pandemic.Research limitations/implications>Despite significant implications, this study has limitations. A response from only two companies is investigated to the COVID-19 pandemic. The scope of this study is only India, a developing nation, thereby, cross country research is recommended. A comparative study between developed and developing countries may be conducted. A quantitative approach may be used to get empirical findings of the COVID-19 pandemic and post-pandemic policies of companies from an international perspective. Hence, there is ample opportunity to research organisations' response to the pandemic and CSR as a strong arm to deal with critical disasters.Practical implications>The paper offers new insights into exploring research and praxis agenda for collaborative potentials towards the evolution of CSR and sustainability.Social implications>The findings develop new initiatives and combat manifold issues of education, employment and hunger during the pandemic to provide quick relief.Originality/value>The paper offers new insights into how companies are considering issues related to the crisis, including avoidance of layoffs and maintaining wage payments, and may be in a better position to access fresh capital, relief programs and emergency funds. Taking proactive health and safety measures may avert legal risks to the company. It is likely that the way in which companies are responding to the crises is a real-life test on resilience and adaptation.